jgeofil / comp551-2

Mini Project 2

Abstract Prediction Text Classification Problem

Assignment for Comp 551, Fall 2016, McGill University

Usage

The entry point of the program is runner.py, where all the applications can be run by these simple flags :

usage: runner.py [-h] [-lr] [-nv] [-sgd] [-rand] [-erf] [-bag]
                 [-vot] [-ada] [-svm]

optional arguments:
  -h, --help            show this help message and exit
  -lr, --logistic       logistic Regression
  -nv, --naive          Naive Bayes
  -sgd, --sgdclassifier
                        SGD Classifier
  -rand, --randomf      Random Forest
  -erf, --extrarf       Extra Random Forest
  -bag, --bagging       Bagging Ensemble
  -vot, --voting        Voting Ensemble
  -ada, --adaboost      Adaboost
  -svm, --svm           Linear SVM

K Nearest Neighbors implementation was done in R so use RStudio to run it.

Authors

• Jeremy Georges-Filteau - McGill ID : 260713547, [email protected]
• Yu Luo - McGill ID : 260605878, [email protected]
• Koustuv Sinha - McGill ID: 260721248, [email protected]